Comment on Janda et al.
There is a great deal of interest in the role that nicotine and its metabolites might play in potential therapies for Alzheimer's disease. This study from Tobin Dickerson and Kim Janda of the Skaggs Institute of Chemical Biology looks in detail at glycation of the amyloid b protein by a nicotine metabolite. The justification for this analysis was given as the existing evidence that nicotine exposure leads to delayed onset of Alzheimer's disease, citing one paper.

This assertion needs addressing. The literature suggesting that there is a protective effect from smoking was based on animal studies and early epidemiological studies, mostly of case-control design. Case-control studies are subject to bias. Longitudinal studies are generally felt to provide a less biased result. The longitudinal studies of smoking and Alzheimer's disease have either produced no association (e.g., Doll et al., 2000), or overall increased risk (e.g.,   Read more

Comment on Janda et al.
There is a great deal of interest in the role that nicotine and its metabolites might play in potential therapies for Alzheimer's disease. This study from Tobin Dickerson and Kim Janda of the Skaggs Institute of Chemical Biology looks in detail at glycation of the amyloid b protein by a nicotine metabolite. The justification for this analysis was given as the existing evidence that nicotine exposure leads to delayed onset of Alzheimer's disease, citing one paper.

This assertion needs addressing. The literature suggesting that there is a protective effect from smoking was based on animal studies and early epidemiological studies, mostly of case-control design. Case-control studies are subject to bias. Longitudinal studies are generally felt to provide a less biased result. The longitudinal studies of smoking and Alzheimer's disease have either produced no association (e.g., Doll et al., 2000), or overall increased risk (e.g., Tyas et al., 2003). The literature has been reviewed systematically by Almeida and colleagues, 2002, and a review of the impact of study design has been published relatively recently (Kukull, 2001).

While laboratory results are of interest as they investigate possible molecular mechanisms, which may or may not bring therapeutic benefit, it is important that the full range of epidemiological evidence is provided for readers, rather than the single paper, which justifies the particular molecular approach.

View all comments by Carol Brayne